Nearly 3 Million Searches of New Hanover’s Flock Cameras

World map displaying global data flow and search statistics from three million computer searches
Seventeen North Carolina communities have already said no to Flock. Add your name ›  ·  The Cape Fear is under warrantless AI surveillance — it’s time to act.
← All updates

Three Million Searches, Zero Names

What the Sheriff’s Office released about its automated license plate reader network — and what it left out.

Key fact — cite this

New Hanover County’s Flock license-plate network was searched 2,980,082 times in about 16 months — and the county redacted the name of every searching agency and officer.

Source: DeFlockILM analysis of New Hanover County Sheriff’s Office audit records (Jan 23, 2025 – Jun 5, 2026), obtained by public-records request. Published at DeFlockILM.org, June 2026.

In April we asked the New Hanover County Sheriff’s Office for the audit records behind its contract with Flock Safety — the automated license plate reader (ALPR) cameras now mounted across the county. The records arrived in June. They cover roughly sixteen and a half months, from January 23, 2025 through June 5, 2026, and they log 2,980,082 searches run through the county’s Flock camera network. Here is what the data shows, and, just as importantly, what it cannot.

The numbers, at a glance
  • 2,980,082 searches in ~16.5 months (Jan 23, 2025 – Jun 5, 2026)
  • 6,177 camera networks touched by the median search; up to 7,216
  • 99.7% were interactive queries typed by a person
  • 100% of organization, officer, and plate names: redacted

A local search that reaches across the country

What this number is — and what it isn’t

The honest version first, because this figure gets misread in both directions and the misreadings are load-bearing.

2,980,082 is not 2.98 million investigations of New Hanover County residents. It is not 2.98 million searches run by New Hanover deputies either. This is a network audit. It logs every query that reached into this county’s cameras, no matter where that query originated. When a detective in Texas runs one broad search across the Flock network, that single search can touch thousands of participating networks at once — and this county’s is one of them. One search there. One row here.

So what does the number measure? Reachability. It is the count of times this county’s camera data was pulled into somebody’s query. And that is precisely the thing worth knowing — because it is not something the county can wave off by noting that most of the searches weren’t its own.

They weren’t its own. That is the finding. A network that this county pays for, that this county’s residents drive past, answering millions of questions asked by people who don’t live here and don’t answer to anyone here.

Which leaves the only question that matters: whose searches were they? These records were produced in a form that cannot answer it. Every organization name is blacked out. We can tell you this county’s data was reached nearly three million times in sixteen months. We cannot tell you by whom — and neither can you, and neither, on this record, can your commissioners.

Each search in the log records how many separate camera networks it touched. The typical search reached a median of 6,177 networks at once; the largest reached 7,216. That is not a glitch. The county’s contract pays for Flock’s State and Nationwide Network lookup — a feature Flock markets as access to “a nationwide system boasting 10 billion additional plate reads per month.” In practice, one officer running one search is querying thousands of other agencies’ cameras across the country, not just New Hanover’s. And this is overwhelmingly human activity: 99.7% of the searches are interactive queries typed by a person, not automated background polling.

Who is doing the searching? The records don’t say.

Every organization name and every officer name in the released log is blacked out — replaced with “***” in all 2.98 million rows. License plates are redacted too. So while the data shows how many searches ran, how far they reached, and what reasons officers typed, it does not show who ran them. We also asked for the lists of agencies New Hanover shares ALPR data with and receives it from, for both license plate data and “hot list” information. Because the names are redacted and no separate lists were provided, those questions remain unanswered. The unredacted version of this log exists — the Sheriff’s Office and Flock hold it — but it was not released.

We know what that unredacted log would look like, because other North Carolina agencies released theirs with the names intact. Appalachian State University runs just three Flock cameras in Boone — and the network audit it produced names all 3,487 agencies that searched them in a single quarter, led by Houston, Texas Police at 177,461 searches, with 84% of the traffic coming from outside North Carolina. New Hanover’s log is the same kind of record. The county simply chose to black out the column App State left readable. We combined five such agencies’ unredacted audits — nearly 3.9 million searches, ~90% from out of state, with North Carolina data then governed by Texas and federal rules (how we counted).

What the searches were for

For the county’s own searches, drug enforcement dominates: 37.4% were coded Drugs/Narcotics, the single largest category by a wide margin. Across the full network log, the picture is murkier. More than half of all searches carry only a generic reason — “Other,” “investigation,” or a blank — and case numbers are missing or redacted in nearly every row. That free-text field is messy, so this is best read as a documentation gap rather than proof that any individual search lacked a good reason. The log also contains 470 “freeform” searches in which officers described a vehicle in plain English — “black Porsche SUV with paper tags,” “white Silverado with a cooler in the bed” — and let Flock’s system hunt for matches by appearance alone. Across all 2.98 million searches, Flock’s content moderation flagged almost nothing: a handful of entries out of millions.

The legal questions this raises

North Carolina has a statute for this. N.C. Gen. Stat. § 20-183.32 makes captured plate data confidential and not a public record, and allows it to be disclosed only to a criminal-justice officer, for a legitimate law-enforcement purpose, and “pursuant to a written request from the requesting agency.” A companion section bars using ALPR data to enforce traffic violations. The Sheriff’s own policy, SOP 222, goes further: it restricts access to office employees and says an outside agency may obtain data only by sending the ALPR manager a written request that names the crime, the case, and the plate — a request the manager must review, approve, and record.

Set that against the data. The records show standing, network-wide access at a scale of nearly three million searches — a structure in which participating agencies can query the county’s cameras without a separate, case-by-case request. Whether a blanket, opt-in network-sharing arrangement satisfies a statute that speaks of “a written request from the requesting agency” is a real and unsettled question. No North Carolina court has yet construed this statute — the question is one of first impression in this state.

To be clear: the released data does not prove that anyone broke the law. Because the identities are redacted, that conclusion is impossible from these records alone. What the data does is surface a genuine conflict between how the system operates and how the law and the county’s own policy say it should — and it points to exactly the records that would settle the matter.

What we’re asking for next

SOP 222 says certain records must exist. We will be asking for them: the ALPR manager’s log of written sharing requests from outside agencies; the annual audits the policy requires be sent to the Sheriff; and the access logs that, by policy, record who viewed the data by name, date, and time. Those documents would answer the question the redacted log cannot — who has been searching New Hanover’s cameras, and whether the process the law requires was followed.

See for yourself

We show our work. The full records the county produced, and our complete analysis of them, are posted below. Read them, check the numbers, and draw your own conclusions — then ask your local leaders why a camera network in Wilmington is fielding millions of searches the public has no way to trace.

Your county commissioners or town council hold the lever that ends Flock where you live. They need to hear from you.
Your move

You’re not a suspect. So stop being tracked like one.

It takes one minute. Add your name — then tell the officials who represent you to take the cameras down.